CN104066058B - A kind of WLAN indoor orientation methods based on double set fingerprint superpositions - Google Patents
A kind of WLAN indoor orientation methods based on double set fingerprint superpositions Download PDFInfo
- Publication number
- CN104066058B CN104066058B CN201410332222.3A CN201410332222A CN104066058B CN 104066058 B CN104066058 B CN 104066058B CN 201410332222 A CN201410332222 A CN 201410332222A CN 104066058 B CN104066058 B CN 104066058B
- Authority
- CN
- China
- Prior art keywords
- point
- fingerprint
- grid
- determined
- coordinate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013519 translation Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000004891 communication Methods 0.000 abstract description 2
- 230000004807 localization Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000000060 site-specific infrared dichroism spectroscopy Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Landscapes
- Collating Specific Patterns (AREA)
Abstract
A kind of WLAN indoor orientation methods based on double set fingerprint superpositions, belong to the technical field of radio communication, are the methods to being positioned to the indoor occupant/assets under radio local network environment.The method gathers the signal characteristic of reference point in off-line phase first, sets up first set RSS fingerprint databases;Then two sets of fingerprints is interlocked and be superimposed all referring to the position coordinates of point in translation first set fingerprint database, then gather the signal characteristic of the reference point after translation, set up second set of RSS fingerprint database.In the tuning on-line stage, grid position of the point to be determined in two sets of fingerprint maps is first judged respectively, take average using grid element center point coordinates and positioned;If local positioning environment is undesirable, can also select optimal a set of fingerprint by the way of screening fingerprint is intersected and be positioned, effectively weaken the influence that indivedual exception fingerprints are caused to location Calculation.The present invention is simple and easy to apply, and positioning precision is high.
Description
Technical field
The invention belongs to the technical field of radio communication, the present invention is fixed in a kind of WLAN rooms based on double set fingerprint superpositions
Position method, it is adaptable to the indoor occupant/assets positioning under radio local network environment.
Background technology
With the rise and fast development of Internet of Things, location Based service application is more and more extensive, location-based clothes
Business requires that alignment system provides accurate, real-time, reliable positional information.Therefore, existing WLAN is made full use of
Believe (Wireless Local Area Networks, WLAN), the accurate position provided with relatively low cost under indoor environment
Breath, has important practical significance.
In current indoor orientation method using it is most common be location fingerprint method, the method is by reference point RP (Reference
Point) signal intensity RSS (the Received Signal measured from wireless access point AP (Access Point)
Strength) match as fingerprint characteristic and geographical space, simple and easy to apply without increasing additional hardware, positioning precision is higher.
Location fingerprint method is divided into two steps of off-line phase and on-line stage, mainly including nearest neighbor method, k-nearest neighbor and probabilistic method.
Wherein k-nearest neighbor (KNN, K Nearest Neighbors) has some superiority on algorithm complex and positioning precision, but
RSS is easily influenceed by multipath, diffraction, diffraction, the personnel disturbing factor such as walk about, and joins the individual candidate calculated by KNN algorithms
Examination point can cause certain influence away from true point to be determined on positioning precision.
Traditional method that fingerprint is covered using foundation list in off-line phase based on RSS location fingerprints positioning mode, indoors
Spaced apart in positioning region to be uniformly arranged reference point and gather RSS, on-line stage needs to go matching in the way of traveling through
Finger print information in location fingerprint map.Positioning precision is lifted with increasing with reference to dot density in theory, due to offline
Stage needs artificial participation to set up fingerprint base, and in order to obtain positioning precision higher, the amount of working offline can also increase therewith, this meeting
Time cost is set to improve rapidly.
Patent name:KNN localization methods, number of patent application in WLAN rooms based on neighbour's points optimization:
201010154412.2, in 2010 times, KNN localization methods in a kind of WLAN rooms based on neighbour's points optimization are disclosed, should
Method is solved in existing WLAN rooms in KNN localization methods, and the positioning precision caused by neighbour's points selection is improper deteriorates
Problem, but the method target area is divided into uniform grid in off-line phase, it is necessary to manually gather each reference
The RSS of point, the amount of working offline is very big.
Patent name:WLAN indoor orientation methods based on partition information entropy production, number of patent application:
201210329662.4, time:2012, a kind of WLAN indoor orientation methods based on partition information entropy production are disclosed, should
Method carries the operand that reduces needed for positioning and by screening AP located space subregion by using K mean cluster algorithm
High position precision, in order to reduce systematic error, raising refers to dot density, but the method divides target area in off-line phase
1 meter of uniform grid is, it is necessary to manually gather the RSS of each reference point, the amount of working offline is very big at interval.
Patent name:Indoor orientation method based on classification thresholds and signal intensity weight, number of patent application:
201310155441.4, in 2013 times, a kind of indoor orientation method based on classification thresholds and signal intensity weight is disclosed,
The method determines matching threshold and using reference point signal intensity as power by according to path loss feature to reference point classification
K nearest neighbor weighting is participated in again, can reduce the influence that indoor environment interference causes RSS randomized jitters, weaken or even elimination is seriously done
The influence of the reference point disturbed, relatively accurately realizes indoor positioning, but be divided into for target area in off-line phase by the method
The uniform grid at 1.5 meters of interval is, it is necessary to manually gather the RSS of each reference point, the amount of working offline is very big.
For above-mentioned background content, a kind of simple and easy to apply, amount of working offline of research is substantially reduced and positioning precision remains to protect
The location fingerprint location algorithm of higher level is demonstrate,proved, it is significant to WLAN indoor positionings.
The content of the invention
The invention aims to improve existing location fingerprint location algorithm, there is provided a kind of to cover what fingerprints were superimposed based on double
WLAN indoor orientation methods, can lift positioning precision, weaken RSS and the disturbing factor such as are walked about by multipath, diffraction, diffraction, personnel
The influence caused to positioning, can substantially reduce the workload that off-line phase fingerprint builds storehouse again, simple and easy to apply, so that indoor to realize
It is accurately positioned with important application value.
The technical scheme that the present invention is provided comprises the following steps:
A, area to be targeted is divided into uniform grid, the length of side of each grid is l, single set common m reference node of fingerprint
Point, by the summit of grid position as a reference point and receives from the n RSS value of AP, structure first set RSS fingerprint maps;
Whole m reference points in B, translation first set fingerprint map, build second set of RSS fingerprint map, and combine two
Set fingerprint map structuring rectangular coordinate system;
C, tuning on-line stage, according to the real-time RSS values of point to be determined, first set fingerprint and second where it are judged respectively
Set fingerprint in position, and using its affiliated grid as the set fingerprint candidate lattices;
D, the candidate lattices selected respectively in two sets of fingerprint maps, the position of point to be determined is determined according to special algorithm.
The step A is:
A1, the k signal intensity RSS values one n × k square of composition from each AP for receiving each reference point
Gust, the ith row and jth column of the matrix is represented from i-th RSS value of AP jth time measurement respectively, wherein, i≤n, j≤k.
A2, each row summation respectively to n × k matrixes of all reference points are averaged again, and its value is stored in
In matrix, orderWherein m represents the sum of reference point, and n is represented
The sum of AP.
The step B is:
B1, first set fingerprint level first translated into downwards l/2 distances to right translation l/2 distances, then level, equally by grid
Summit position as a reference point and receive from the n RSS value of AP, during its value existed into signal intensity RSS matrixes, specifically
Method is shown in A1.
B2, build second set of fingerprint mapMatrix, specific method is shown in A2.
B3, using second set of fingerprint bottom as x-axis, it is positive direction that orientation is right, and the Far Left of first set fingerprint is made
It is y-axis, is positive direction in orientation, respectively by the center point coordinate of each grid in two sets of fingerprints and its position in fingerprint map
Put numbering one-to-one corresponding to be stored in two different GC matrixes, makeWherein
I represents the row sum of the grid of reference in fingerprint map, and j represents the row sum of the grid of reference in fingerprint map, and (x, y) represents grid
Center point coordinate.
The coordinate of each reference point in two sets of fingerprints is stored in two different RP matrixes respectively, is madeWherein i represents the row sum of reference point in fingerprint map, and j represents fingerprint
The row sum of reference point in map, (x, y) represents the coordinate of reference point.
The step C is:
C1, the real-time RSS sampling numbers of point to be determined are N, are averaged real-time RSS of the signal strength values as the point
Value, calculates each reference point in real-time RSS and two set of fingerprint map respectivelyEuclidean distance between value, using following public affairs
Formula:
Wherein, DiRepresent real-time RSS to DiEuclidean distance value between individual reference point, i=1,2 ... ..., m, m are reference
Point sum, j=1,2 ... ..., n, n are AP sums.
In C2, a set of fingerprint wherein, preceding four Euclidean distances D values of minimum are found, and find its corresponding reference
Point coordinates, first takes out the reference point coordinates corresponding to first three minimum Euclidean distance, judges any two of which reference point
Whether horizontal, the difference DELTA x of ordinate and Δ y meets regional determination condition, and (0≤Δ x≤0≤Δs of 2l ∩ y≤2l, l represent grid
The length of side).
If meeting above-mentioned condition, weight index is calculated by equation below:
Wherein, i=1,2,3;K=1,2,3.
Average coordinates of the point to be determined in the set fingerprint are calculated by equation below:
Wherein, i=1,2,3.
Judge which grid point to be determined belongs in fingerprint map, use equation below:
Wherein, dijThe distance of point to be determined and point coordinates in arbitrary mess in fingerprint map is represented,Represent to be positioned
Coordinate of the point in the set fingerprint map, (xij,yij) represent grid element center point coordinates, i=1,2 ... ..., j=in fingerprint map
1,2 ... ....
Choose minimum dijThe corresponding coordinate of value is the grid element center point coordinates in fingerprint map where point to be determined,
Candidate lattices using the grid as point to be determined in the set fingerprint.
If C3, being unsatisfactory for above-mentioned condition, the reference point coordinates corresponding to preceding four Euclidean distance D values of minimum is chosen,
This four average coordinates of reference point are calculated, is represented with following formula:
The distance between this four reference points and its average coordinates value dis is calculated respectivelyi, represented with following formula:
By disiIn maximum corresponding to coordinate weed out, using remaining three coordinate values elder generation Rule of judgment, if
Meet condition, then calculate weight index ωi, calculate point to be determined coordinate and find its grid in fingerprint map, specific step
Suddenly C2 is seen;
If being still unsatisfactory for condition, then it is assumed that point to be determined cannot be by reference to a side for cluster in the set fingerprint map
Method is accurately positioned, and position error is larger, i.e., real grid position cannot judge, now chooses point to be determined on set fingerprint ground
Coordinate in figure corresponding to minimum the first two Euclidean distance value.
The step D is:
If there are candidate lattices in D1, two sets of fingerprint maps, it is averaged with the center point coordinate of two candidate lattices
Value is used as point to be determined coordinate.
If only existing a candidate lattices in D2, two sets of fingerprint maps, with first three minimum for judging grid establishment
Coordinate corresponding to Euclidean distance point is averaged as point to be determined coordinate.
If not existing candidate lattices in D3, two sets of fingerprint maps, take nearest with point to be determined in respective fingerprint map
The first two reference point, calculate this four average values of point as point to be determined coordinate.
Beneficial effects of the present invention:
(1) using double set fingerprint map Fold additon locations, point to be determined region area can be reduced four in theory/
Three, average localization error declines, and positioning precision is lifted;
(2) the use of side length of element is double set fingerprint map Fold additon locations of l, is single set fingerprint of l/2 compared to side length of element
With suitable positioning precision, the quantity that can but make off-line phase choose reference point declines 50%, greatlys save time cost;
(3) under complicated indoor conditions, if local positioning environment is undesirable, can also be by the way of screening fingerprint be intersected
Select optimal a set of fingerprint to be positioned, effectively weaken the influence that indivedual exception fingerprints are caused to location Calculation.
Brief description of the drawings
The double set fingerprint superposition algorithm schematic flow sheets of Fig. 1.
Experimental situation panorama schematic diagram in Fig. 2 examples.
The double set fingerprint superposition schematic diagrames of Fig. 3.
Specific embodiment
The present invention is elaborated with reference to specific embodiments and the drawings, the implementing platform of the embodiment is Windows XP
Operating system, signal intensity is gathered using the wireless network card Wireless-N1030 of Dell notebook computer InspironN4110,
Double set fingerprint Fold additon location algorithm flow schematic diagrames are shown in Fig. 1.
A, area to be targeted is divided into uniform grid, if the length of side of each grid is l=1.2m, totally 24 reference nodes
Point, by the summit of grid position as a reference point and receives from 6 RSS values of AP, structure first set RSS fingerprint maps,
Experiment panorama is shown in Fig. 2, wherein, yellow net region is positioning region, and experiment sets up 6 AP altogether, and SSID name is respectively
MMCL1-6, is operated in 1,6,11 these three different channels, and SSID name is for DLUT for AP is had in place by oneself, and all experiments are equal
In 27.6 × 1.8m2Carried out in region;
A1,50 signal intensity RSS values one 6 × 50 squares of composition from each AP for receiving each reference point
Battle array, sample frequency is 1 time/second, during its value existed into signal intensity RSS matrixes.
A2, each row summation respectively to 6 × 50 matrixes of all reference points are averaged again, and its value is stored in
In matrix.
All 24 reference points in B, translation first set fingerprint map, build second set of RSS fingerprint map, and combine two
Set fingerprint map structuring rectangular coordinate system, is shown in Fig. 3;
B1, first set fingerprint level first translated into downwards 0.6m distances to right translation 0.6m, then level, equally by grid
Summit position as a reference point is simultaneously received from 6 RSS values of AP, during its value existed into signal intensity RSS matrixes.
B2, build second set of fingerprint mapMatrix.
B3, using second set of fingerprint bottom as x-axis, it is positive direction that orientation is right, and the Far Left of first set fingerprint is made
It is y-axis, is positive direction in orientation, the center point coordinate of each grid in two sets of fingerprints is stored in two different GC respectively respectively
In matrix.
The coordinate of each reference point in two sets of fingerprints is stored in two different RP matrixes respectively, in two sets of fingerprint maps
Particular location of the Position Number of each reference point in rectangular coordinate system is shown in Fig. 3.
C, tuning on-line stage, according to the real-time RSS values of point to be determined, first set fingerprint and second where it are judged respectively
Set fingerprint in position, and using its affiliated grid as the set fingerprint candidate lattices;
C1, the coordinate of point to be determined are (x, y), and real-time RSS sampling numbers are 5 times, are averaged signal strength values conduct
The real-time RSS values of the point, each reference point in real-time RSS and two set of fingerprint map is calculated using formula (1) respectivelyValue
Between Euclidean distance.
C2, in first set fingerprint, find minimum first three Euclidean distance D1、D2And D3Value, and find its corresponding ginseng
Examination point coordinate (x1,y1)、(x2,y2) and (x3,y3), judge that these three sit with reference to the horizontal, vertical of any two reference point in point coordinates
Whether target difference DELTA x and Δ y meet regional determination condition (0≤Δ x≤0≤Δs of 2l ∩ y≤2l, l represent the length of side of grid).
If meeting above-mentioned condition, the weight index of these three reference points is calculated respectively by formula (2):ω1、ω2With
ω3, and bring these three weighted values average coordinates of formula (3) the calculating point to be determined in the set fingerprint intoThen lead to
The distance that formula (4) calculates each grid element center point coordinates in the average coordinates and fingerprint map respectively is crossed, is finally chosen most
Grid element center point coordinates corresponding to small distance point and the candidate lattices using the grid as point to be determined in the set fingerprint.
If C3, being unsatisfactory for above-mentioned zone decision condition, preceding four Euclidean distance D of minimum are chosen1、D2、D3And D4Value institute
Corresponding reference point coordinates (x1,y1)、(x2,y2)、(x3,y3) and (x4,y4), calculate the flat of this four reference points using formula (5)
Equal coordinateThen the average coordinates are substituted into formula (6), calculate respectively this four reference points and its average coordinates value it
Between apart from disi, by disiIn maximum corresponding to coordinate weed out, first judge bar using remaining three coordinate value
Part, if meeting regional determination condition, calculates weight index ωi, determine point to be determined coordinate and find it in fingerprint map
Grid, specific steps are shown in C2;
If being still unsatisfactory for regional determination condition, then it is assumed that point to be determined cannot be by reference to point in the set fingerprint map
The method of cluster is accurately positioned, and position error is larger, i.e., real grid position cannot judge, now chooses point to be determined at this
Coordinate in set fingerprint map corresponding to minimum the first two Euclidean distance value.
D, the candidate lattices selected respectively in two sets of fingerprint maps, the position of point to be determined is determined according to special algorithm.
If there are candidate lattices in D1, two sets of fingerprint maps, it is averaged with the center point coordinate of two candidate lattices
Value is used as point to be determined coordinate.
If only existing a candidate lattices in D2, two sets of fingerprint maps, with first three minimum for judging grid establishment
Coordinate corresponding to Euclidean distance point is averaged as point to be determined coordinate.
If not existing candidate lattices in D3, two sets of fingerprint maps, take nearest with point to be determined in respective fingerprint map
The first two reference point, calculate this four average values of point as point to be determined coordinate.
Under this experimental situation, using the present invention, the indoor average localization error for obtaining is 0.95m.
Claims (1)
1. it is a kind of based on double WLAN indoor orientation methods for covering fingerprints superposition, it is characterised in that following steps:
A, area to be targeted is divided into uniform grid, the length of side of each grid is l, single set common m reference mode of fingerprint, general
The summit of grid position as a reference point is simultaneously received from the n signal intensity RSS value of reference point AP, structure first set RSS
Fingerprint map;
Whole m reference points in B, translation first set fingerprint map, build second set of RSS fingerprint map, and refer to reference to two sets
Line map structuring rectangular coordinate system;It is specific as follows:
(1) first set fingerprint level is first translated into downwards l/2 distances to right translation l/2 distances, then level, equally by the top of grid
Point position as a reference point and reception from the n RSS value of AP, during its value existed into signal intensity RSS matrixes;
(2) second set of fingerprint map is builtMatrix, order
Wherein m represents the sum of reference point, and n represents the sum of AP;
(3) using second set of fingerprint bottom as x-axis, it is positive direction that orientation is right, using the Far Left of first set fingerprint as y
Axle, is positive direction in orientation, respectively by the center point coordinate of each grid in two sets of fingerprints and its position in fingerprint map
Numbering is corresponded and is stored in two different GC matrixes, is madeWherein i
The row sum of the grid of reference in fingerprint map is represented, j represents the row sum of the grid of reference in fingerprint map, and (x, y) represents grid
Center point coordinate;
The coordinate of each reference point in two sets of fingerprints is stored in two different RP matrixes respectively, is madeWherein, i represents the row sum of reference point in fingerprint map, and j is represented and referred to
The row sum of reference point in line map, (x, y) represents the coordinate of reference point;
C, tuning on-line stage, according to the real-time RSS values of point to be determined, judge first set fingerprint where it respectively and second set refers to
Grid position in line, and using its affiliated grid as the set fingerprint candidate lattices;
It is specific as follows:
(1) the real-time RSS sampling numbers of point to be determined are N, are averaged real-time RSS value of the signal strength values as the point, point
Each reference point in real-time RSS and two set of fingerprint map is not calculatedEuclidean distance between value, using equation below:
Wherein, DiRepresent real-time RSS to DiEuclidean distance value between individual reference point, i=1,2 ... ..., m, m are that reference point is total
Number, j=1,2 ... ..., n, n are AP sums;
(2) in a set of fingerprint wherein, preceding four Euclidean distances D values of minimum are found, and finds its corresponding reference point and sat
Mark, first takes out the reference point coordinates corresponding to first three minimum Euclidean distance, judges the horizontal, vertical of any two of which reference point
Whether the difference DELTA x and Δ y of coordinate meet 0≤Δ of regional determination condition x≤0≤Δs of 2l ∩ y≤2l, and l represents the length of side of grid;
If meeting above-mentioned condition, weight index is calculated by equation below:
Wherein, i=1,2,3;K=1,2,3;
Average coordinates of the point to be determined in the set fingerprint are calculated by equation below:
Wherein, i=1,2,3;
Judge which grid point to be determined belongs in fingerprint map, use equation below:
Wherein, dijThe distance of point to be determined and point coordinates in arbitrary mess in fingerprint map is represented,Represent that point to be determined exists
Coordinate in the set fingerprint map, (xij,yij) represent fingerprint map in grid element center point coordinates, i=1,2 ... ..., j=1,
2 ... ...;
Choose minimum dijThe corresponding coordinate of value is the grid element center point coordinates in fingerprint map where point to be determined, by this
Candidate lattices of the grid as point to be determined in the set fingerprint;
(3) if being unsatisfactory for above-mentioned condition, the corresponding reference point coordinates of preceding four Euclidean distance D values of minimum is chosen, is calculated
This four average coordinates of reference point, are represented with following formula:
The distance between this four reference points and its average coordinates value dis is calculated respectivelyi, represented with following formula:
By disiIn maximum corresponding to coordinate weed out, using remaining three coordinate values elder generation Rule of judgment, if meeting bar
Part, then calculate weight index ωi, calculating point to be determined coordinate and find its grid in fingerprint map, specific steps are shown in C
(2);
If being still unsatisfactory for condition, then it is assumed that point to be determined cannot be by reference to the method essence of a cluster in the set fingerprint map
It is determined that position, position error is larger, i.e., real grid position cannot judge, now chooses point to be determined in the set fingerprint map
Coordinate corresponding to minimum the first two Euclidean distance value;
D, the candidate lattices selected respectively in two sets of fingerprint maps, the position of point to be determined is determined according to special algorithm;Specifically such as
Under:
(1) if there are candidate lattices in two sets of fingerprint maps, averaged work with the center point coordinate of two candidate lattices
It is point to be determined coordinate;
(2) if only existing a candidate lattices in two sets of fingerprint maps, with first three the minimum Euclidean for judging grid establishment
Coordinate corresponding to range points is averaged as point to be determined coordinate;
(3) if not existing candidate lattices in two sets of fingerprint maps, take it is nearest with point to be determined in respective fingerprint map before
Two reference points, calculate this four average values of point as point to be determined coordinate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410332222.3A CN104066058B (en) | 2014-07-14 | 2014-07-14 | A kind of WLAN indoor orientation methods based on double set fingerprint superpositions |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410332222.3A CN104066058B (en) | 2014-07-14 | 2014-07-14 | A kind of WLAN indoor orientation methods based on double set fingerprint superpositions |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104066058A CN104066058A (en) | 2014-09-24 |
CN104066058B true CN104066058B (en) | 2017-07-11 |
Family
ID=51553548
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410332222.3A Active CN104066058B (en) | 2014-07-14 | 2014-07-14 | A kind of WLAN indoor orientation methods based on double set fingerprint superpositions |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104066058B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109581286A (en) * | 2018-12-28 | 2019-04-05 | 皖西学院 | A kind of two dimension fingerprint distribution indoor positioning method and structure |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105554762B (en) * | 2015-12-10 | 2019-01-04 | 广东工业大学 | Wireless spoofing attack localization method based on RSS |
CN105916113B (en) * | 2016-06-01 | 2019-04-12 | 北京邮电大学 | A kind of localization method and device |
CN105916201A (en) * | 2016-06-07 | 2016-08-31 | 天津大学 | KNN fingerprint positioning method based on RSS Euclidean distance fitting |
CN106970351B (en) * | 2017-05-15 | 2019-09-06 | 河海大学常州校区 | WLAN fingerprint rescan localization method based on signal propagation model |
CN108737979B (en) * | 2018-05-11 | 2020-05-22 | 华南理工大学 | Indoor positioning method |
CN108693518B (en) * | 2018-05-15 | 2020-11-27 | 西北大学 | Indoor positioning method |
CN109089313B (en) * | 2018-09-12 | 2020-10-09 | 河南迈驰物联网有限公司 | Double-fingerprint joint positioning method and device |
CN110738143B (en) * | 2019-09-27 | 2023-06-02 | Oppo广东移动通信有限公司 | Positioning method and device, equipment and storage medium |
CN111426316B (en) * | 2020-06-15 | 2020-09-25 | 北京云迹科技有限公司 | Robot positioning method and device, robot and readable storage medium |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102880673A (en) * | 2012-09-10 | 2013-01-16 | 电子科技大学 | Indoor positioning method |
-
2014
- 2014-07-14 CN CN201410332222.3A patent/CN104066058B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102880673A (en) * | 2012-09-10 | 2013-01-16 | 电子科技大学 | Indoor positioning method |
Non-Patent Citations (3)
Title |
---|
"Adaptive Temporal Radio Maps for Indoor Location Estimation";Jie Yin, Qiang Yang, Lionel Ni;《Third IEEE International Conference on Pervasive Computing and Communications》;20050321;全文 * |
"Dynamic Online-Calibrated Radio Maps for Indoor Positioning in Wireless Local Area Networks";Mohamed M. Atia,Aboelmagd Noureldin,Michael J. Korenberg;《IEEE Transactions on Mobile Computing》;20120626;全文 * |
"基于k-近邻算法的定位方法设计和仿真";李文杰,李文明;《计算机仿真》;20090415;全文 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109581286A (en) * | 2018-12-28 | 2019-04-05 | 皖西学院 | A kind of two dimension fingerprint distribution indoor positioning method and structure |
CN109581286B (en) * | 2018-12-28 | 2020-12-04 | 皖西学院 | Two-dimensional fingerprint distribution indoor positioning method and structure |
Also Published As
Publication number | Publication date |
---|---|
CN104066058A (en) | 2014-09-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104066058B (en) | A kind of WLAN indoor orientation methods based on double set fingerprint superpositions | |
CN106792465B (en) | A kind of indoor fingerprint map constructing method based on crowdsourcing fingerprint | |
CN103747419B (en) | A kind of indoor orientation method based on signal strength difference and dynamic linear interpolation | |
CN103476118B (en) | A kind of WLAN indoor location fingerprint positioning method for monitoring in real time | |
CN101873607B (en) | WLAN (Wireless Local Area Network) indoor step-type RD-ANFIS (Region Division-Adaptive Network-based Fuzzy Inference System) positioning method | |
CN108696932A (en) | It is a kind of using CSI multipaths and the outdoor fingerprint positioning method of machine learning | |
CN104038901B (en) | Indoor positioning method for reducing fingerprint data acquisition workload | |
CN104936148A (en) | Indoor positioning method for WIFI (Wireless Fidelity) based on fuzzy KNN (k-Nearest Neighbor) | |
CN103209478A (en) | Indoor positioning method based on classified thresholds and signal strength weight | |
CN106535326A (en) | WiFi locating method based on depth neural network and server | |
CN104076327B (en) | Continuous positioning method based on search space reduction | |
CN102395194B (en) | ANFIS (Adaptive Neural Fuzzy Inference System) indoor positioning method based on improved GA(Genetic Algorithm) optimization in WLAN (Wireless Local Area Network) environment | |
CN106507476A (en) | A kind of WiFi localization methods and server and location model construction method | |
CN109068267A (en) | A kind of indoor orientation method based on LoRa SX1280 | |
CN106792540A (en) | A kind of improvement DV Hop localization methods based on route matching | |
Jiang et al. | Wi-Fi fingerprint based indoor localization without indoor space measurement | |
CN108882363A (en) | A kind of multi-direction acquisition combines the WiFi fingerprint indoor orientation method of cluster | |
CN104581945B (en) | The WLAN indoor orientation methods of semi-supervised APC clustering algorithms based on distance restraint | |
CN106597370A (en) | Indoor wireless positioning method based on RFID | |
Hashimoto et al. | SICNN: Spatial interpolation with convolutional neural networks for radio environment mapping | |
CN104581644A (en) | Indoor WLAN fingerprint database multipoint self-adaptive updating method based on radial basis interpolation | |
Wu et al. | Cooperative motion parameter estimation using RSS measurements in robotic sensor networks | |
Cheng et al. | An Anchor Node Selection Scheme for Improving RSS‐Based Localization in Wireless Sensor Network | |
Sadhukhan et al. | A novel weighted fusion based efficient clustering for improved wi-fi fingerprint indoor positioning | |
CN105872987B (en) | A kind of Distributed Wireless Sensor Networks node positioning method based on single mobile robot |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |